Upload 3 files
Browse files- README(7).md +46 -0
- config(1).json +24 -0
- generation_config(1).json +6 -0
README(7).md
ADDED
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
pipeline_tag: text-generation
|
4 |
+
language:
|
5 |
+
- en
|
6 |
+
tags:
|
7 |
+
- pretrained
|
8 |
+
inference:
|
9 |
+
parameters:
|
10 |
+
temperature: 0.7
|
11 |
+
---
|
12 |
+
|
13 |
+
# Model Card for Mistral-7B-v0.1
|
14 |
+
|
15 |
+
The Mistral-7B-v0.1 Large Language Model (LLM) is a pretrained generative text model with 7 billion parameters.
|
16 |
+
Mistral-7B-v0.1 outperforms Llama 2 13B on all benchmarks we tested.
|
17 |
+
|
18 |
+
For full details of this model please read our [paper](https://arxiv.org/abs/2310.06825) and [release blog post](https://mistral.ai/news/announcing-mistral-7b/).
|
19 |
+
|
20 |
+
## Model Architecture
|
21 |
+
|
22 |
+
Mistral-7B-v0.1 is a transformer model, with the following architecture choices:
|
23 |
+
- Grouped-Query Attention
|
24 |
+
- Sliding-Window Attention
|
25 |
+
- Byte-fallback BPE tokenizer
|
26 |
+
|
27 |
+
## Troubleshooting
|
28 |
+
|
29 |
+
- If you see the following error:
|
30 |
+
```
|
31 |
+
KeyError: 'mistral'
|
32 |
+
```
|
33 |
+
- Or:
|
34 |
+
```
|
35 |
+
NotImplementedError: Cannot copy out of meta tensor; no data!
|
36 |
+
```
|
37 |
+
|
38 |
+
Ensure you are utilizing a stable version of Transformers, 4.34.0 or newer.
|
39 |
+
|
40 |
+
## Notice
|
41 |
+
|
42 |
+
Mistral 7B is a pretrained base model and therefore does not have any moderation mechanisms.
|
43 |
+
|
44 |
+
## The Mistral AI Team
|
45 |
+
|
46 |
+
Albert Jiang, Alexandre Sablayrolles, Arthur Mensch, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Florian Bressand, Gianna Lengyel, Guillaume Lample, Lélio Renard Lavaud, Lucile Saulnier, Marie-Anne Lachaux, Pierre Stock, Teven Le Scao, Thibaut Lavril, Thomas Wang, Timothée Lacroix, William El Sayed.
|
config(1).json
ADDED
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"MistralForCausalLM"
|
4 |
+
],
|
5 |
+
"bos_token_id": 1,
|
6 |
+
"eos_token_id": 2,
|
7 |
+
"hidden_act": "silu",
|
8 |
+
"hidden_size": 4096,
|
9 |
+
"initializer_range": 0.02,
|
10 |
+
"intermediate_size": 14336,
|
11 |
+
"max_position_embeddings": 32768,
|
12 |
+
"model_type": "mistral",
|
13 |
+
"num_attention_heads": 32,
|
14 |
+
"num_hidden_layers": 32,
|
15 |
+
"num_key_value_heads": 8,
|
16 |
+
"rms_norm_eps": 1e-05,
|
17 |
+
"rope_theta": 10000.0,
|
18 |
+
"sliding_window": 4096,
|
19 |
+
"tie_word_embeddings": false,
|
20 |
+
"torch_dtype": "bfloat16",
|
21 |
+
"transformers_version": "4.34.0.dev0",
|
22 |
+
"use_cache": true,
|
23 |
+
"vocab_size": 32000
|
24 |
+
}
|
generation_config(1).json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 1,
|
4 |
+
"eos_token_id": 2,
|
5 |
+
"transformers_version": "4.34.0.dev0"
|
6 |
+
}
|